374 BUFFON 



they were to be painted on the under side of the 

 labels.^ 



Then Buffon sketches such a natural history as he 

 would himself approve. Knowledge is divisible into 

 Civil History and Natural History.^ Natural History 

 should mainly consist of descriptions of natural objects, 

 not minute, not formal ; they should above all be read- 

 able. The grouping should be such as would suggest 

 itself to the first human observer, and the succession 

 should be determined by the closeness of the relation to 

 man. It is a mistake to let the description of the zebra, 

 which is foreign and unfamiliar, follow that of so well- 

 known and useful an animal as the horse ; the dog, 

 which we are accustomed to see running at the horse's 

 heels, might much more fitly come in this place. The 

 natural and ordinary way of looking at things is the 

 best. 



As he continued his descriptions Bufibn discovered 

 that all this was impracticable. In his chapter on the 

 Degeneration of Animals he stated his views on the 

 arrangement of the quadrupeds, without a hint that he 

 had ever discussed the question before, and put forth as 

 his own a classification which is substantially Kay's.' 

 Every useful group which it contains is taken from 

 Ray's Synopsis Methodica Animalium Quadrupedum. 

 Ray's mistakes are left uncorrected, but Bufibn adds a 

 few of his own. The Cetacea are still excluded from 

 the Mammalia; the hippopotamus is still separated from 

 the swine ; BufFon cannot tell, any more than Ray, 



^ The story goes that Linnseus revenged himself upon Buffon by naming 

 the toad Bufonia, which he never did. 



^ This division is much older than Buffon ; it occurs for instance in Bacon's 

 De Augmentia. 



' It is not, however, to be forgotten that he had given an account of Ray's 

 classification of quadrupeds in his fourth volume {Discours sur la nature des 

 animaux). 



